DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The response filed 11/24/2025 has been entered. Applicant has not amended, added, or cancelled any claims. Claims 1-20 are currently pending in the instant application.
Response to Arguments
Applicant's arguments filed 11/24/2025 have been fully considered but they are not persuasive. Regarding the arguments concerning the independent claims, the Examiner respectfully disagrees. Claim 1 recites:
A computer-implemented method for generating a query, the method comprising: receiving a user input; selecting, from a plurality of predefined inputs, at least one predefined input based on similarity to the user input; and prompting a first trained machine learning model to generate a query based on a prompt that comprises the user input and at least one predefined query associated with the at least one predefined input.
Hemington teaches the claimed “prompt” as further explained in [0058] that details the inputs into the LLM disclosed in Hemington.
[0058] Inputs to an LLM may be referred to as a prompt, which is a natural language input that includes instructions to the LLM to generate a desired output. A computing system may generate a prompt that is provided as input to the LLM via its API. As described above, the prompt may optionally be processed or pre-processed into a token sequence prior to being provided as input to the LLM via its API. A prompt can include one or more examples of the desired output, which provides the LLM with additional information to enable the LLM to better generate output according to the desired output. Additionally, or alternatively, the examples included in a prompt may provide inputs (e.g., example inputs) corresponding to/as may be expected to result in the desired outputs provided. A one-shot prompt refers to a prompt that includes one example, and a few-shot prompt refers to a prompt that includes multiple examples. A prompt that includes no examples may be referred to as a zero-shot prompt.
The prompt disclosed in Hemington contains both the user input ([0058] - the natural language input to the LLM) and at least one predefined query associated with the at least one predefined input ([0058] - Additionally, or alternatively, the examples included in a prompt may provide inputs (e.g., example inputs) corresponding to/as may be expected to result in the desired outputs provided). Applicant further narrows the “prompting” limitation in dependent claim 7 and 16 where the predefined input is an example input, which is the same language that is disclosed in Hemington. For these reasons, the Examiner maintains the current rejection.
Examiner recommends further clarifying amendments to claim 1 to recite language from Figure 4, 410-418, specifically the conditions required to either execute the query, prompt the model for a corrected query, or display the query and errors to the user, to receive a corrected query from the user. For example: “in response to a unsuccessful query execution and a current attempt is less than a maximum number of attempts, prompt the language model to generate a new query that corrects error(s) produced during query execution.”.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by HEMINGTON et al (US20240320251).
HEMINGTON discloses:
1.A computer-implemented method for generating a query, the method comprising: receiving a user input ([0012] In an aspect, the present application discloses a computer-implemented method. The method may include: obtaining at least one query; clustering a set comprising the at least one query into first clusters; for each first cluster, identifying, by a large language model (LLM), queries in the cluster that are semantically dissimilar; clustering the queries identified as semantically dissimilar into one or more second clusters; receiving an incoming query; matching the incoming query to a particular cluster from the first or second clusters; obtaining one or more generated response messages based on providing, to the LLM, data associated with the particular cluster for the incoming query. ); selecting, from a plurality of predefined inputs, at least one predefined input based on similarity to the user input ([0012] In an aspect, the present application discloses a computer-implemented method. The method may include: obtaining at least one query; clustering a set comprising the at least one query into first clusters; for each first cluster, identifying, by a large language model (LLM), queries in the cluster that are semantically dissimilar; clustering the queries identified as semantically dissimilar into one or more second clusters; receiving an incoming query; matching the incoming query to a particular cluster from the first or second clusters; obtaining one or more generated response messages based on providing, to the LLM, data associated with the particular cluster for the incoming query. ); and prompting a first trained machine learning model to generate a query based on a prompt that comprises the user input and at least one predefined query associated with the at least one predefined input ([0012] In an aspect, the present application discloses a computer-implemented method. The method may include: obtaining at least one query; clustering a set comprising the at least one query into first clusters; for each first cluster, identifying, by a large language model (LLM), queries in the cluster that are semantically dissimilar; clustering the queries identified as semantically dissimilar into one or more second clusters; receiving an incoming query; matching the incoming query to a particular cluster from the first or second clusters; obtaining one or more generated response messages based on providing, to the LLM, data associated with the particular cluster for the incoming query. )
2. The computer-implemented method of claim 1, wherein selecting the at least one predefined input comprises: generating an embedding of the user input; computing a plurality of distances between the embedding of the user input and a plurality of embeddings of the plurality of predefined inputs; and selecting the at least one predefined input based on the plurality of distances and a distance threshold ([0071] The query processing engine 114 is configured to compute similarity between the vectors in an embedding space. In particular, the query processing engine 114 may use one or more metrics for calculating vector similarity such as, but not limited to, L2 (Euclidean) distance, cosine similarity, and inner product (dot product). Various algorithms for vector similarity search may be implemented by the search engine. Examples include k-nearest neighbor (kNN), approximate nearest neighbors (ANN) search, space partition tree and graph (SPTAG), Faiss, and hierarchical navigable small world (HNSW).[0072] The clustering module 118 may perform clustering using the vector embeddings that are generated by the embedding module 116. In particular, the clustering module 118 may identify clusters in the embedding space. Clustering operations may be performed by implementing a suitable cluster model (e.g., connectivity model, centroid model, etc.) and clustering algorithm (e.g., DBSCAN, agglomerative clustering, spectral clustering, etc.). The clustering module 118 is configured to output information regarding clustering operations such as, for example, cluster labels, clustering algorithms, distance metric(s), linkage criterion, and cluster membership.).
3. The computer-implemented method of claim 1, further comprising: generating the plurality of predefined inputs based on one or more database fields and one or more first predefined templates; and generating a plurality of predefined queries associated with the plurality of predefined inputs based on the one or more database fields and one or more second predefined templates ([0077] A query management system may build and maintain a database of queries and corresponding responses. More particularly, as will be described in greater detail below, the database may include sets of questions/issues that are identified from query messages received by the system and one or more solutions corresponding to each question/issue. The query database may be built, for example, by processing large datasets of previously received queries. The system may use the query database in generating responses to incoming (and future) queries. Additionally, the system may dynamically update the database based on new query data.).
4. The computer-implemented method of claim 3, wherein the plurality of predefined inputs are further generated based on one or more user inputs [0066] The query processing engine 114 is configured to receive user-supplied queries from user devices 120 via a network 150. In the context of a service platform, the user-supplied queries may be customer queries. A customer query may comprise an email message, a portal message, a chat input, etc. submitted by a customer that contains at least one question or issue.
5. The computer-implemented method of claim 1, further comprising: displaying the query to a user; receiving, from the user, a corrected query that corrects the query; and storing the user input as a new predefined input and the corrected query as a new predefined query ([0093] In some implementations, a correct solution to an extracted query may be dependent on metadata associated with the querying user of the platform (e.g., a merchant) such as, for example, location of the merchant's store, apps installed on the merchant's store, subscription plan, history of the store or merchant, or other state information associated with the merchant. The feature set for the query embeddings may additionally include these parameters relating to the metadata (or other representations of contextual data).).
6. The computer-implemented method of claim 1, further comprising determining one or more key terms included in the user input, wherein the first trained machine learning model is further prompted to generate the query based on at least one of a database field or a formula associated with the one or more key terms [0091] The feature set corresponding to the feature vectors may include two parts. The first part of the feature set may relate to the text of the extracted query itself. That is, embeddings may be generated from the text of the extracted query. The text embeddings may be generated, for example, using an OpenAI™ embeddings call. As explained above, domain-specific information may provide additional context that enables an LLM to identify granular distinctions between queries of a same cluster.
7. The computer-implemented method of claim 1, wherein prompting the first trained machine learning model to generate the query comprises inputting, into the first trained machine learning model, a prompt that includes the user input and the at least one predefined query as at least one example[0099] For each query and solution combination, the computing system may supply the LLM with the query, the solution (and all associated “solution steps”), and any additional resources (e.g., text of help documentation), and instruct it to generate a solution paragraph (operation 408). The final output dataset produced by the LLM may contain a set of queries and a defined number of potential solution paragraphs per query. The output may be used to update a query and response database, in operation 410. In particular, the computing system may add the output dataset to a database storing queries and responses data. [0104] The methods and systems described herein may be deployed in part or in whole through a machine that executes computer software on a server, cloud server, client, firewall, gateway, hub, router, or other such computer and/or networking hardware. The software program may be associated with a server that may include a file server, print server, domain server, internet server, intranet server and other variants such as secondary server, host server, distributed server and the like. The server may include one or more of memories, processors, computer readable media, storage media, ports (physical and virtual), communication devices, and interfaces capable of accessing other servers, clients, machines, and devices through a wired or a wireless medium, and the like. The methods, programs or codes as described herein and elsewhere may be executed by the server. In addition, other devices required for execution of methods as described in this application may be considered as a part of the infrastructure associated with the server.
8. The computer-implemented method of claim 1, further comprising: executing the query using a database system; and in response to receiving an error from the database system, prompting a second trained machine learning model to generate a corrected query based on the user input, the query, and the error [0095] The computing system may then instruct the LLM to verify whether the queries in a given cluster are indeed the same. More particularly, the LLM may be used to “refine” the clusters of the embeddings (operation 310). Queries in a cluster that are the same query may be labelled with a new query that best describes all the “same queries”. In this way, new labels may effectively be generated at cluster level. The queries that are different may be separated out into their own clusters. That is, the LLM may be instructed to create a new label (or keep them unlabeled) for queries that are deemed to be dissimilar to other queries within the cluster. An advantage of using the LLM in this step is that it is able to distinguish between queries in the same cluster that may have met the defined threshold criteria (i.e., vectors that are within a threshold distance from each other) but that may be substantially different semantically.
9. The computer-implemented method of claim 8, wherein the first trained machine learning model is the second trained machine learning model [0041] This process of instructing the LLM to generate a query corresponding to the first data request based on modifying an input prompt to the LLM may proceed iteratively until a successful response is received from the endpoint. Upon determining a successful query, i.e., a query that is accepted by the endpoint, the system may update a knowledge base, such as a queries database, storing query information of queries for the endpoint. The knowledge base may indicate previous queries that were accepted by the endpoint and/or examples of queries that would be accepted, i.e., queries that are constructed to comply with requirements of the endpoint and a specific query language. The system may add a successful query to the knowledge base to include a record associated with said query. In particular, the successful query may be stored, in a database, in association with an indication of the first data request..
10. The computer-implemented method of claim 1, wherein the first trained machine learning model comprises a large language model (LLM). [0041] This process of instructing the LLM to generate a query corresponding to the first data request based on modifying an input prompt to the LLM may proceed iteratively until a successful response is received from the endpoint. Upon determining a successful query, i.e., a query that is accepted by the endpoint, the system may update a knowledge base, such as a queries database, storing query information of queries for the endpoint. The knowledge base may indicate previous queries that were accepted by the endpoint and/or examples of queries that would be accepted, i.e., queries that are constructed to comply with requirements of the endpoint and a specific query language. The system may add a successful query to the knowledge base to include a record associated with said query. In particular, the successful query may be stored, in a database, in association with an indication of the first data request.
Claims 11-20 are rejected using similar reasoning seen in the rejection of claims 1-10 above due to reciting similar limitations but directed towards a non-transitory computer readable media and a system.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SAMUEL SHARPLESS whose telephone number is (571)272-1521. The examiner can normally be reached M-F 7:30 AM- 3:30 PM (ET).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, ALEKSANDR KERZHNER can be reached at 571-270-1760. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/S.C.S./Examiner, Art Unit 2165
/ALEKSANDR KERZHNER/Supervisory Patent Examiner, Art Unit 2165